Skip to content

release prep: v0.2.0 - #101

Merged
open-coder-ai merged 2 commits into
mainfrom
release/v0.2.0-prep
Sep 1, 2026
Merged

release prep: v0.2.0#101
open-coder-ai merged 2 commits into
mainfrom
release/v0.2.0-prep

Conversation

@open-coder-ai

Copy link
Copy Markdown
Owner

What this changes

The four version-bump edits that precede a release, nothing else: pyproject.toml, src/agentseam/__init__.py, CITATION.cff0.2.0, and CHANGELOG.md gets a [0.2.0] - 2026-09-01 heading over the accumulated Unreleased entries (D2–D6, ACS vocabulary, fixes). No tag is created by this PR — merging it publishes nothing; the release happens only when the owner creates the v0.2.0 tag.

Why 0.2.0, not the earlier-planned 0.1.2

This release renames the canonical Decision vocabulary (askescalate, rewritetransform, with deprecated aliases kept), changes the documented degraded_from evidence value to its canonical spelling (a called-out break in #95), adds the transform capability keys, and re-expresses eleven adapters as engine + vendor config. Under 0.x SemVer convention that's a minor bump; calling it a patch would under-signal. If the owner prefers 0.1.2, the change is these same four lines.

Pre-verified (dry run, nothing published)

  • Wheel built from main and twine check PASSED; cold-venv install: CLI smoke green, all 13 data/vendors/ files ship, bundles render at the D6 sizes, an unfilled HANDLER template fails closed.
  • Full suite on this branch: 1399 passed, 4 skipped; the version-consistency tests (pyproject/CITATION track __version__) pass with the bump.
  • Known downstream note for after the release: chock's pin bump is not mechanical — its evidence.py validates claim wire-words against the canonical vocabulary, which splits in this release. Findings and the fix plan are in org-plan plan/spine-a/pin-bump-findings.md; the fix lands with the pin bump, and chock stays green on 0.1.1 until then.

Claim check

Not applicable — no adapter, matrix row, or capability claim is touched; version metadata only.

Checks

  • pytest -q passes (1399 passed, 4 skipped)
  • ruff check . and ruff format --check . unaffected (no Python logic touched)
  • Runtime path still stdlib-only (no code changes)
  • Commit signed off

🤖 Generated with Claude Code

https://claude.ai/code/session_01RcphZmVVXoy8Jf2tUtcsC3


Generated by Claude Code

Version 0.2.0 rather than the earlier-planned 0.1.2: this release renames the
canonical Decision vocabulary (ask->escalate, rewrite->transform, aliases kept),
changes the documented degraded_from evidence value to its canonical spelling,
adds the transform capability keys, and re-expresses eleven adapters as engine
plus vendor config. Under 0.x SemVer convention that is a minor bump, not a
patch. No tag is created here; tagging and publishing are the release owner's
action.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Claude <noreply@anthropic.com>
@open-coder-ai
open-coder-ai marked this pull request as ready for review September 1, 2026 21:41
The social-preview card renders the package version, so the bump invalidated
the committed derivation; regenerated with docs/assets/gen_brand_assets.py
(the required brand-assets check's own remedy).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Signed-off-by: Claude <noreply@anthropic.com>
@open-coder-ai
open-coder-ai merged commit acb3881 into main Sep 1, 2026
14 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants